Valve Undo Some Steam User Review Changes, A Bit

Valve have reversed some of their recent changes to Steam user reviews. Most notably, they’ve made store pages once again include all reviews, no matter how players obtained the game. Some developers we’d spoken to were concerned that excluding people who had got games by activating Steam keys – by backing Kickstarters, buying in other stores, and so on – could harm them by making their games seem worse and less popular. But some devs were glad for the change, an attempt by Valve to crack down on fake reviews. Well, now all reviews are once again shown by default, though overall ‘scores’ will ignore reviews from key copies.

Valve’s changes last week cracked down on developers using Steam keys (which they get for free) to fuel fraudulent positive reviews, which boosted their scores, visibility, and sales. Valve explained that “it’s becoming increasingly difficult to detect when this is happening, which reviews from Steam Keys are legitimate, and which are artificially influenced.” So they made a grand sweeping change: all reviews from keys on all games would be, by default, filtered out from all the places reviews are shown and tabulated.

Goodbye fraudulent reviews (hooray!), and goodbye reviews from people who got keys by backing Kickstarters, buying games on Itch or the Humble Store, in Humble Bundles, and so on (oh). In short, some devs feared they were losing positive reviews by people who genuinely played and enjoyed their games.

Given that Steam is the biggest PC gaming shop and hub, problems for developers are problems for all of us – people who like playing neat games. Fiddling with visibility can make some good games harder to discover and therefore play. How it might affect developers financially is an issue too. Without devs we don’t have games, and many of my favourite games have come from small developers who are heavily reliant upon Steam and Valve. This might only be one change but it’s a fragile ecosystem, and we should be concerned.

Valve have now reversed part of the change. By default, Steam store pages once again include all reviews, whether players bought the game directly through Steam or got a key elsewhere then activated it on Steam. However, the aggregate user review score will only include reviews from copies bought directly through Steam.

Steam’s ‘scores’ aren’t numbers but rather ratings derived from the percentage of reviews which give a game a ‘Recommended’ thumbs-up or a ‘Not Recommended’ thumbs-down. These ratings – which are Overwhelmingly Negative, Mostly Negative, Negative, Mixed, Positive, Mostly Positive, and Overwhelmingly Positive – are displayed at the top of Store pages and in listings.

That means this is still a problem. Reviews are nice, but review scores are more prominent and more likely to be used in snap judgments. This is better but not ideal. Phoney reviews caused problems, absolutely, but so does this.

As Paul Kilduff-Taylor of Mode7 told us last week, “It’s frustrating that the idiotic behaviour of a few unscrupulous people has, once again, forced changes to a system which affects everyone.”

On the brighter side, Valve’s announcement last night did also talk about filtering out the wakka wakka wacky reviews that some pranksters boost ‘for the lulls’. They said:

“Our existing system just looks at the overall number of users that rated a review as ‘helpful’, but we’re seeing this can produce unpredictable results. For example, sometimes unhelpful memes get rated as ‘helpful’ because people think it’s funny. So we’re working on updating the system to consider more factors when deciding how to rank ‘helpful’ reviews so that it can generate better results.”

Valve have changed the colour game pages use to portray a Mixed score too. It’s gone from a dark orange (which had a pretty negative implication) to a neutral yellow matching the icon used in search results. Good-o.


  1. snv says:

    To improve the system, they should just get rid of these review scores.

    Often the actual review would lead me not buy the game, but still it is marked as a buy recommendation, and sometimes in reviews marked as negative the actual flaw descibed would be something i don’t mind.

    Now how much someone cares for a specific flaw or feature is subjective and thats why the attempt to boil it down to a total average is just wrong.

    Though considering that RPS is against scores, this seems to be a bit of preaching to the choir.

    • HeavyStorm says:

      Well spoken.

    • Czrly says:

      Actually… exactly this! Great idea.

    • LTK says:

      The scores are a natural consequence of collecting a large number of ‘Recommended’/’Not Recommended’ ratings in one place. Naturally someone who’s doing research into a game will want to know how many people recommend it versus how many don’t; it’s a valid statistical measure, unlike the scores that many websites assign to their reviews that always carry a lot of arbitrariness.

      Removing the score will solve nothing, because the reviews are publicly available and if Valve will not aggregate them into an overall score, someone else will. Whether or not scores are an accurate measure of a game’s quality is questionable, but as a representation of the distribution of the reviews, I don’t think they could have picked a better measure.

      • LionsPhil says:

        Removing the thumb-up/thumb-down would.

        Admittedly, that’d take away the every-useful “most helpful negative” search, i.e. “what flaws does this game have (and are they ones that’d ruin it for me)”.

      • Premium User Badge

        Phasma Felis says:

        The majority of buyers will go by the score listed on Steam. People who wait and check Metacritic before buying are a minority.

    • SuicideKing says:

      I would disagree, I don’t have time to read everyone’s subjective opinion on a game.

      May be better to use a 5 or 10 point scale instead of just 0/1? Alternatively – recommended, qualified recommendation, not recommended.

    • BlazeHedgehog says:

      Could use a system like Google does on the Play store, where it calculates a score AND looks for common keywords to highlight repeated praises or complaints.

    • Ericusson says:

      Hell no.

      • Ericusson says:

        For a longer than a Hell no answer, it’s pretty much like baseball and stats. Everybody in Baseball was playing like old farts relying on an undefinable gut feeling and experience BS until stats, you know crunching numbers, were used and reality was uncovered about real players performance under the above said layer of BS.

        Percentile notes in Steam rely on a market with enough people to give feedback so as to be statistically significative (with limits for smaller titles).

        Besides, I want to have a quick view of the game quality, and I have it with these magical things called numbers, if I am interested but not convinced, detailed reviews are there then.

        As for RPS, I enjoy having no notes on this website, but it is not Steam, the store where I buy games.
        The notes are actually one of the key elements of the success of STEAM in my views, relying on a form of social network of the buyers to share their experience of the product.

        If buyers want to rate a game on steam when they buy it on another platform, then get a socialist president and congress in the states and nationalize Steam for being so
        Dominant on the market to open it to anyone (ok this one is not really argumented).

        So yeah, Hell No.

        • Ericusson says:

          The argument against notes by one website is that they are too subjective, the arguments for notes given by 10 000 people is that they are statistically significative.

          • snv says:

            It’s the other way around.

            Even if those notes were aggregated by absolutely everyone, they’re pointless because you try to generate an objective measure for something which only has a subjective value.

            The average might be interesting for marketing people and investors and so on, but for the individual gamer it’s mostly useless.

            They might help if you are the most average person on earth, but the more your personal taste differs from the average, the less useful the average opinion becomes.

            The other aspect is the oversimplification. In your baseball example you have a very homogeneous matter — every baseball game works with the same mechanics as any other, every pitcher has to do the same things as any other pitcher and so on — and still you have hundreds of meaningful metrics there.

            (Computer) Games though are very heterogeneous. In every genre you have different aspects, many games defy being put into a genre, and (to repeat the first argument) even in each genre every potential player assigns different importance to different features.
            And then this very multidimensional amorphous topic is being boiled done to _one_ number. The madness!

          • Ericusson says:

            SNV, so according to you one website review is objective and a 10 000 aggregated grades is subjective.

            Thinking is good but you sometimes need to make sure you keep your feet on the ground when you process leads you to assert such an obviously wrong conclusion.

            It feels you don’t understand what are statistics.

          • snv says:

            That is not what i meant. Because that notion is ridiculous, you could have assumed that this was not what i meant.

            Statistics is a powerful tool, but that does not help when it is used at the wrong problem.

            When the question for a potential buyer is “Will I like that game?” statistics is useless, if it is used this way.

            The other way around, like those amazon recommendations for example, it can help, but that way focuses on the individuals preferences instead of the agglomerate.

          • April March says:

            Reviews are always subjective. The average score of ten thousand reviews is still subjective, because each review score is subjective. The only objective data you can get from that is how well liked the game is, which for me personally says nothing – many of my favourite games have been hated by almost everyone, and many fan-favourites I found dreadful.

            Conversely, while a review score is subjective, a single review may state objective facts. If I read a review that says “this stupid game has horrible red cars” I have just been given the objective information that the game has red cars, which I might or might not like. I used a deliberately stupid thing, but for a real example I once read a review that panned Resident Evil 4 for not having a lot of backtracking; that would count as a major point in favour of the game for me (if I hadn’t already played the game years before I saw that review).

    • Sizeable Dirk says:

      The avergage is a great tool to quickly dismiss the utterly broken sub 40% shovelware flooding Steam and spend browsing-time actually reading the reviews of more promising titles.

    • geldonyetich says:

      I’m on board with the idea of ditching a percentage scoring of thumbs up/thumbs down. Frankly, user reviews are most often bandwagons to overly positive or overly negative. A surprising lot of Steam reviews are ironic, since the players have no reason to take doing their reviews seriously. Ditch the scores, make it so people have to read the reviews, the review system gets much, much better.

    • Chaoslord AJ says:

      I got no problem at all with scores and find them immensely useful in making buying decisions.
      Maybe they should have a side thumb but in all honesty if a game’s considered “side-thumed” others probably shouldn’t play it as the good games already in existence exceed your life time to play thoroughly. No need to play 0-70% rating games.
      I have yet to see an example where the rating keeps players from enjoying a “great” game rather than saving their precious time.
      For example I enjoyed Mankind Divided and it’s only “mixed” (technical issues, in-game store, ending and so on) but if you skip it you’re not missing out either.

    • wengart says:

      Steam reviews let you, at a glance, get an idea of the relative quality of a game.

      I never read Steam reviews. I only care about them when they reach 2,000 reviews or more. At which point the pure number of reviews means that there is generally some amount of truth in the number.

    • falconne says:

      No, getting rid of the scores is throwing the baby out with the bathwater. The review system isn’t perfect and never will be, but if it wasn’t for Steam reviews the industry will go back to that awful Metacritic rating obsession.

      The vast majority of people want a score and if they don’t get it here, they will go to Metacritic and AAA companies will swamp out the indie devs again.

  2. Czrly says:

    Not impressed. I thought the changes were for the good, mostly. I will be interested to see what they do about the meme-reviews and rep- and neg-rep- mobs.

  3. aircool says:

    I wanted to write a positive review for Elder Scrolls Online: Gold Edition as I think it’s worth every penny. Unfortunately, Steam wouldn’t let me.

    The Steam review system is in a strange place at the moment. Deus Ex: MD is a good case in point. On the one hand, a lot of the negative reviews weren’t about the game (which I enjoyed on my 70+ hour playthrough), but about either bugs and poor optimisation and the day one DLC.

    Whilst none of that actually reflects on the game, it does have some validity. If you can’t get the game working, or at least working to be playable on the minimum specs, then you’re not going to be happy and recommend the game to anyone else.

    Likewise, Day One DLC and Season Pass bollocks. Many people don’t like it. Those two things alone just make you feel like you’ve bought a full price game, but with parts of it behind another paywall, and that’s not a good feeling.

    When publishers are looking at so many negative reviews which don’t reflect the gameplay, they need to take listen to their customers. Shitty console ports and poorly optimised games, along with DLC/Season Pass crap aren’t popular amongst gamers, which, lets face it, isn’t anything that we haven’t heard before.

    The Steam reviews are, unfortunately for the publishers, doing what they’re supposed to be doing. If your game is getting kicked, there’s usually a really good reason for it.

    • Hedgeclipper says:

      Indeed, I’m far from convinced by the argument in the article that developer’s interests are aligned with consumers here. Similarly skeptical that reviews from kickstarters who’ve been following and invested in a game form the beginning are likely to point out things that might concern me looking for a finished polished product as a purchase.

      • Nerdy Suit says:

        The interests of the devs and consumers are not aligned at all. Frankly, the tone that these RPS articles are taking on this subject is ridiculous and come across as anti-consumer.

        Sorry but…NO…the devs problems are not at all my problems. Don’t say that they are. My problem is to find a good game that I would enjoy to spend my money on. The devs job is to sell that game. How in the hell are our interests aligned??? Actually, they’re interests are really the exact opposite of mine. They’re trying to sell me something and I’m trying to make sure I buy the right thing. The devs interests are no more aligned with my interests than the car salesman who’s desperately trying to sell any car they can at the absolute highest price they can. Actually saying that the devs problems are my problems is not only factually incorrect, but also anti-consumer and acting as a shill for the devs, who at the end of the day are just another business.

        The resolution to this is pretty simple: These devs should stop worrying about who is and isn’t getting filtered and worry much more about making a good game with minimal bugs that will likely get good reviews from most consumers, not just a certain population that got their keys in a particular way. I mean, if your KS backers overall give your game a 90% and everyone else who purchase via Steam gave it a 70%…shouldn’t you be asking yourself why? Why the huge disparity? Perhaps the game isn’t as good as you think it is? Perhaps your KS backers are biased for obvious reasons and don’t see the valid flaws?

        • RobF says:

          It’s weird to me to see ‘our interests aren’t aligned’, this seems like a really small view of how everything works and kinda unfairly antagonistic to most people in and around games.

          One of the larger problems with the review system changes is that they effect discovery. This is absolutely not just a problem for developers, it’s for customers too. (also why abusing the system as either a developer or a customer is ultimately short sighted idiocy). It’s everyone’s problem.

          Steam currently ties in reviews and review percentages, amongst other things, to how much a game gets surfaced and what slots it gets in the store. This is not so simple as ‘just make a good game and it’ll sort itself out’, that plainly just doesn’t happen. We have years and years of data and shared knowledge to know that it does not work this way. So.

          What does happen is that good games fall between the cracks and don’t sell, or some sell to the wrong people (like people who would never have an interest in them) and plenty of excellent games just don’t get found or picked up by the public. It’s not self regulating or self correcting, it’s an enormous problem.

          So what this leads to is developers who make work that might well appeal to one segment of the audience can’t reach that audience and that audience can’t find the games they would like. You can see this writ large in the way the app store works. So both developers and customers need discovery to work.

          Most of the concerns around the review system are around making sure people who want the games can find the games. In an increasingly noisy market, this is an enormously difficult issue to work around because there isn’t like a gestalt entity of people who all like the same thing. It’s also incredibly precarious, a slight change in one system can ripple and effect developers and customers *clicks fingers* like that.

          Now, I appreciate no-one has to concern themselves with this if they just want to nip onto a store and buy videogames but someone has to concern themselves with it so that people can nip onto a store and buy the videogames they want to play. As it’s our livelihoods and the actual real future of Steam as a viable platform at stake, of course developers take an interest in it (and some will be flat out idiots, this is par for the course with anything). Because the end game here is that if Steam doesn’t stay a viable platform to sell on, well, no-one puts their games on Steam and boof! Guess we didn’t want it anyway or something.

          • Hedgeclipper says:

            While I don’t necessarily disagree with what you’re saying here a lot of the arguments being thrown about by developers seem contrary to this. Gaming reviews or having your ardent kickstarter supporters push up the score on steam is absolutely adding noise that doesn’t benefit a consumer on steam. In your blog post further down you mention giving games to friends and again positive reviews from your friends and your mum don’t help me as a consumer. More over the more developers who do this the more it damages the discoverability of other games by shunting them down the rankings.

            Anyway this isn’t about discovering ‘good’ games, its about discovering -popular- games. I’m don’t think anyone has a reliable method for even working out what a ‘good’ game is given that can mean so many things to different people.

          • Baines says:

            But the interests of consumers and the interests of developers are not aligned. Nor are the interests of Steam aligned with either of those.

            Consumers want to buy games that they will like, and want the best value for their money. Developers want to sell their titles to consumers. As for Valve, Valve wants to sell what will make it money.

            Consumers want to buy “the best games”. Developers want consumers to buy specifically their games, which may or may not be “the best games”. I’m not saying developers are used car salesmen. I’m sure many developers believe in the games that they provide. But just because a developer believes their game is great doesn’t mean it is great. It is human nature to over-value your creations, and we’ve seen plenty of times where developers seem amazed that their games never caught on, seemingly oblivious to “obvious” (at least to consumers) reasons for said titles to fail. (Yes, some developers want good games in general to sell well. That is them drifting into “consumer” mode.)

            And back to Valve, Valve isn’t in line with either. Valve doesn’t care whether or not specific games sell (other than their own, where they are in “developer” mode). At the same time, Valve isn’t necessarily interested in “the best games” selling. Valve would prefer to sell a $60 title that will make them $18 than to sell a $1 title that will make them $0.30. Valve would rather keep a mass appeal game on the front page than a low appeal niche game, because that mass appeal game will sell to more customers and make Valve more money.

            So you have three different groups, all with different interests.

    • Chaoslord AJ says:

      “Technically busted” affects the gameplay. People should again begin to expect working games on release.
      Paying customers are not payed beta testers.
      No one’s obliged to fix a busted release so that’s all the game you get which must be reviewed accordingly. Later you can adjust the review if you want.

  4. DThor says:

    I don’t think anything involving “user reviews” can ever be perfect,and at least they’re trying to make them better. Quite frankly the exceptionally fussy gaming community needs to calm down and just use common bloody sense. You already have a product that you can return after trying with no reason other than “I don’t like it”, reams of “professional” reviews that merely require a bit of experience getting used to the ones that tend to be crap or objective, and yes those hundreds of user reviews. You don’t have to get off your butt to return it, merely find a window to try it out. How many other products out there offer this?

  5. Premium User Badge

    Neurotic says:

    I like the colour change. I tend to see that dark orange colour before I even see the word ‘Mixed’. :D

  6. haldolium says:

    It’s an awfully flawed system to begin with, their changes don’t mean much in the big picture and I found it in general rather ridiculous. The review system has so many issues and the one they go into, creating once again controversity, is the most irrelevant one from a consumer perspective.

    They should be focusing on diverse options instead, trying to actually create a proper representation of the games issues. That would even help developers.

    The amount of ******** people that don’t know how their machine works and go then complain about issues, giving a negative review, became disgustingly high in the past 2-3 years.

  7. Turkey says:

    Why is there an option to mark reviews as funny again?

    • Sleepy Will says:

      It’s to help us spot those who are working on their edinburgh show, and can therefore ignore them as a work of critical review while embracing the idea that this person has chosen a strange and inconvienient way to test their material

    • Xiyng says:

      It’s there so that people would rather mark reviews as funny than useful, if the review in case is funny but not useful.

      • LTK says:

        Exactly this. The internet has conditioned us to go for the ‘like’ or ‘upvote’ button whenever we read something funny, and in Steam’s case the closest thing is the ‘helpful’ button. The ‘funny’ button is supposed to divert that impulse but I doubt it’s effective.

      • LionsPhil says:


        And God forbid there be any fun around games. We must be po-faced and serious in our objective analysis of their merit.

    • aircool says:

      Because those are often the most interesting to read, and can often be obliquely insightful.

    • Baines says:

      It is because quite a while back people complained to Steam that “unhelpful memes get rated as ‘helpful’ because people think it’s funny“.

      People were asking for solutions like a system that would allow users to flag reviews. Basically, solutions that would create more work for Valve.

      Valve instead added a “Funny” button. The idea was apparently that with an option of labeling a review “funny”, people would stop upvoting joke reviews.

  8. mavrik says:

    This essentially also means that your opinion doesn’t matter if you buy a DRM-free version. Nice!

    • Sleepy Will says:

      To be fair though, the reviews on Game’s website also exclude the opinion of those who don’t like purple branding. It’s hardly a new idea that stores only allow reviews from their customers – in Steams case, its consumers of products of customers of their DRM

    • LTK says:

      This is the kind of hyperbolic sentiment that Valve is trying to address. The key activation reviews are now no longer hidden away, so the only thing you lose if you review the game with a non-Steam copy is the ability to contribute to the score.

      If you ask me, if you are at all proficient in writing an informative review, that will probably carry far more weight than the drop in the bucket that your ‘Recommended’ rating will add to the score.

      • Kaeoschassis says:

        I confess I haven’t been following this thing as closely as I should’ve, but wasn’t it mainly smaller-time devs that were upset about the original changes? I was under the impression that one of the major complaints was that for games with smaller audiences to begin with, excluding key-based customers’ scores from the total would have a pretty significant impact, since that represented a big percentage of their total player-base. Remember that steam’s score system doesn’t select its score ‘keyword’ (positive, mostly positive etc. ) based solely on the percentage of positive reviews, but also on the total number; even if 100% of your reviews are positive it doesn’t automatically mean you get an “overwhelmingly positive” score and so on. While the argument can CERTAINLY be made that steam have no obligation or incentive to care about people who don’t buy from them, I definitely feel that the devs themselves are perfectly entitled to be upset if they perceive that steam, through whom they sell their games, is actively hurting their sales. Of course, then you could also argue that if “a significant percentage of their playerbase” didn’t buy through steam, then they aren’t standing to lose as much if less steam users are swayed into a purchase by the game’s overall score, and that’s actually a fair point now I think it over, but ultimately, no dev wants to give up good publicity for their game, and that makes perfect sense, right?

        That was a stream-of-consciousness kind of thing, so forgive the wordiness of it, but to summarise: no, in the case of the people who’re actually raising concerns about the original changes in the first place, any given customer’s review not being considered in the overall score is not “a drop in the bucket”.

    • Premium User Badge

      phuzz says:

      What stops you from reviewing a game that doesn’t have DRM? I’ve been able to leave a review for FTL, and that doesn’t have DRM.

    • Urthman says:

      Why do you think your opinion should matter to Valve if you’re not their customer?

  9. Vacuity729 says:

    I think there’s a fundamental problem here in that there’s an untenable conflict between how the different sides see the purpose of Steam reviews.

    Developers and publishers see them as bonus marketing. They want their most loyal customers to talk about how amazing their product (game) is. They want more reviews, better reviews to lift their product(s) up through the thousands of competing products to the top of the various lists Steam offers. Some of them also want to be able to remove reviews where customers are clearly upset about one or more things regarding the product.

    Consumers want to be able to tell whether a game has technical problems. They want to know if a game is unfinished, or abandoned with extant bugs. They want to know if the developer or publisher is behaving unreasonably (this may be real or invented). And they want to know about the game itself. And, this being the modern Internet, consumers want to be able to give some kind of wrist-slap when a game, developer or publisher fails to meet the expectations they had when they purchased the game.

    Valve is trying to find some kind of balance between these quite different sets of wants, and I honestly don’t think it’s terribly likely that anyone can find that balance. Better to redefine (or provide a reality check to one group of people) what these reviews are really for; marketing, or peer-level communication about the product. Pick one and make that work.

    • Chaoslord AJ says:

      It’s primarily a tool for weeding out the trash. Many people make their living with games. I understand that. They release stuff which they believe to be good and honest work but is usually called shovelware by gamers because it’s derivative and uninspired. Both point of views are true in a way.
      That leads to 95% of trash releases on steam with gamers sifting for the hidden gem all the time. They call it a living, but for me it’s clicking the “not interested” button (which doesn’t seem to do anything).
      Reviews aren’t a tool of advertisement. I don’t usually consider reading a review if the game has not been referenced in high praise in another media or the studio/series has an excellent reputation.
      That’s like Rimworld coverage on RPS and Youtube, that’s Edmund McMillen’s games, that’s CDProjects upcoming cyberpunk, that’s next Elder Scrolls.

  10. jusplathemus says:

    To be honest, I always take the reviews with a grain of salt, because it doesn’t seem like a really good idea to spend my money based on the written words of other people that I don’t even know.
    From the developers’ perpective though, it seems much more complicated, I can see both sides. I do think that Valve is trying here, but there seems to be no easy solution. There will always be dissidence where opinions are allowed.

    • Sizeable Dirk says:

      If a game is good enough there is usually a ton of info through the store even if dismissing user reviews.

      You’ve got the store page presentation with videos, screenshots, the curators’ lists, recommendations from friends if they own it, positioning by popularity in the store lists, a metacritic aggregate with links to professional reviews, sometimes a few award badges, the community pages (bleh) and then the rest of the internet a click away.

      Unless the seller is actively trying to hide something and as long as the customer isn’t just pressing the Early Access Pre-Order button when spotting the Survival and Zombies tags that is.

      • Sizeable Dirk says:

        Oh and Enhanced Steam and you’ll have all the statistics you can possibly chuck that average score against if in doubt.

      • jusplathemus says:

        I was only trying to say that I think it’s unwise to rely solely on Steam reviews. Like you wrote, there are tons of other sources where prople can get more info about a game. I prefer to try it out myself (play a demo if possible) or at least watch a gameplay video.
        But in the end, it’s all subjective, so it doesn’t matter all that much to me what anyone says. There are ill-received games which I adore and conversely, there are games which the majority of the gaming community seems to love that I’m not quite fond of.

  11. Frank says:

    This is exactly the compromise I’d hoped they’d make. Somewhat better would be letting the user save their filter settings.

  12. The Sombrero Kid says:

    Seems to me like a deliberate over reaction + planned minor capitulation to make it look like they are listening and/or caring.

    It saddens me when I look at what valve has become – A pioneer of 21st century corporate responsibility shirking.

    The core issue remains unaddressed, valve are still exploiting their dominant position in the market to damage competition.

  13. SuicideKing says:

    Still can’t see overall most helpful reviews, just most recent ones…

  14. Nerdy Suit says:

    Like I said before, maybe the devs complaining about this should concentrate less on which reviews are/aren’t filtered and concentrate more on making a good game with minimal bugs that most players, regardless of how they got the key, will review positively.

    These changes help consumers filter reviews to their desire to make a more informed purchase decision. The filter is huge, on the main page, completely out in the open, and very easy for any user to see and change according to their desire.

    To me, this is yet another example that the Internet will complain about anything, no matter how good or bad, because that’s what the Internet does.

    Good job, Valve. Thanks for taking this step in an attempt to make the reviews a better tool for me to use to make a more informed purchase decision. I apologize on behalf of the impossible-to-please internet.

    • RobF says:

      If we lived in a world where ‘just making a good game’ and making it as bug free as possible were a path to success and everything working out, sure.

      As it is, I’m dead happy that the changes work for you but us developers do, really, have to consider how these things impact us. Because y’know, without us making the games there isn’t any games and we’d kinda like to be able to carry on where possible.

      • Nerdy Suit says:

        Please, stop using lame straw man arguments. If you don’t make games then games don’t exist? Ummm, ok? That’s true for every industry? That’s true for car manufacturers, but that doesn’t mean that a car manufacturer and its car salesman have the same interests that I do when I go to buy a car. Your stance on this subject is completely ignorant and jaded by the fact that you are a developer. My goal is to buy the game that I would enjoy the most for the cheapest price possible. Your goal is to sell me your game specifically for the highest price possible. That’s just the reality.

        Honestly, you could learn a lesson in “control what you can control and stop worrying about everything else.” The way Steam is set up right now benefits consumers, unquestionably, because it provides them with a greater ability filter reviews how they see fit, which helps them be a more informed consumer according to the type of feedback the consumer is looking for. I don’t want your KS backers saying how great your game is because, honestly, many of them have vested interest in the game succeeding and are probably jaded to some of its flaws.

        Sorry the review system doesn’t benefit you, but honestly, it’s not supposed to benefit you. It’s not supposed to be an extension of your marketing arm. It’s supposed to benefit the consumer, not necessarily the developer. Devs trying to control how reviews are done and filtered is conflict of interest.

        • jusplathemus says:

          This is quite harsh, but I believe you’ve got a point.

  15. genoforprez says:

    Still think there that developers (and game news sites) are mostly seeing this and discussing this as an issue that affects developers and their business. But there is very little discussion of these changes from the consumer perspective. We’re getting news story after news story of posted tweets from developers opining about what this means for indie dev, but nobody seems to care too much what actual Steam users think. It’s like the people on the business side of games are living in their own little bubble.

    It’s like, on one hand we acknowledge that this is an ecosystem, but we’re discussing it as if what concerns developers is the only thing that matters in that ecosystem.

    • RobF says:

      I had a blast at breaking down why this is all happening the other day and more to the point, what it means for Steam as an ecosystem: link to

      I don’t think folks really grasp what Valve are trying to solve for the most part, which is how it all ends up in this mess from all sides.

    • davec1 says:

      I disagree with this assessment. My perception is that the developers do recognize the consumers’ legitimate interest in weeding out fradulent reviews AND actually have a legitimate, aligned interest in that, too, probably a much more significant one, even ( So you now have to get a refund for a game because you were misled by a review? boo-frigging-hoo. For honest devs, people cheating to get more visibility than them threatens their livelihood, and they can’t get anything like a refund for that).

      It’s just that devs are also aware of the unintended consequences of the measures that Valve has taken, while many consumers remain in an ignorant mindset of “it’s not a problem for me, so it’s not a problem” or actually don’t even fully grasp what the issues are to begin with.

      Let’s be clear, the unintended (giving Valve the benefit of the doubt here) consequences harm honest developers, WITHOUT helping consumers. There is no conflict of interest there. There’s only people saying “hey, this is bad for us” and other people saying “why should we care?”. It can even be argued that consumers, too, have an interest in good games with little marketing power getting a chance at some visibility, even if there is an abundance of good games, anyway.

      One thing Valve should at the very least do is get rid of the arbitrary thresholds of number of reviews required to get certain scores (50 & 500). Good algorithms have been proposed for that. I can’t imagine that would even cost them more than their simplistic ones.

      Beyond that, there should be a way to track whether a Steam Key was sold (in which case its owner’s review should count for the score/rank), or whether it was free ( in which case it’s ok to ignore it, even if a free key by no means automatically means an unbalanced review. Else we could all shut down RPS, Kotaku, Eurogamer et al right now). But this is something that I recognize is much harder to implement. For the simplistic algorithm, however, there is no excuse that I’m aware of.

  16. Baines says:

    Valve could just display both overall values, the purchased through Steam and purchase+key. It isn’t like they don’t show outside scores, as store pages include Metacritic scores when available. Heck, it isn’t like the store page doesn’t include that specific information, it is just buried behind the filter buttons at the bottom of the page. (The customer reviews section includes the aggregate of your current filter selection, and you can mouse-over the colored text to see the actual percentage.)

    The bigger problem for me is the time filter for displayed reviews, which users aren’t allowed to change. For many games, that means that even the most helpful reviews won’t be displayed if they are more than a month old. For really popular games, the time filter is a single day, nearly guaranteeing that the “most helpful” reviews that are actually displayed are hardly helpful at all.

  17. Wisq says:

    Regarding excluding reviews from non-Steam-sales … I am okay with this. Steam is a storefront. If I’m looking at a game on Steam, I want to know “should I buy this on Steam?” And the fairest way to answer that question is to see reviews exclusively from other people who bought it on Steam.

    If someone Kickstarted the game, then they were evidently willing to throw money at it back when it was just a concept. I don’t have anything against that — I Kickstart a lot of games myself — but I’m not really sure I’d 100% trust their opinion. I feel like there’s a good chance it may be unduly positive (because it’s “their baby”), or unduly negative (because it didn’t live up to what they thought it would be) … and that’s on top of the fact that they almost certainly didn’t pay the same amount as is being asked for on Steam.

    So yeah, I’m okay with excluding their reviews, especially if it means getting rid of the fraudsters. There’s plenty of other reviewers to choose from.

    Regarding the “this review is funny” button … I see what they’re trying to do, in theory — divert the people who only click “helpful” because it was funny. I’ve seen game devs online who are worried that it just encourages people to write funny reviews — turning devs’ livelihoods into a joke contest — but honestly, that was happening anyway, even without the funny button.

    Frankly, if they wanted to address that (ensuring that reviews are serious), the funny button should just be a silent downvote, or a review should automatically go away if it gets a certain amount of funny votes. I realise that makes me sound like I’m against fun — not at all. But yeah, there are devs depending on thes game sales, and turning everything into a joke is not the best way to support that. (It’s the same reason that, back when I watched Zero Punctuation, I would never have used one of his videos as an actual recommendation — it’s all just for laughs.)

  18. fish99 says:

    Can’t Steam keep track of which keys they’re giving to developers and not allow reviews on those keys? Seems like a better solution than just not counting reviews from all redeemed keys. I’m sure most of my Steam games weren’t bought through Steam.

    • April March says:

      I don’t think there’s any difference Steam-side between keys given to devs (and then potentially given to shady reviewers) and keys generated by Humble et al.

  19. Hobbes says:

    Any solution that requires Valve to do actual -work- is out of the window. Keep this in mind.

    • MadTinkerer says:

      Yeah, but the more time Valve has to spend on community stuff, the less time they get to spend on making the products they want. Valve doesn’t have a department for community management.

      Basically, I’m saying fraudulent reviews are the reason we don’t have Half Life 3 yet. ;)

  20. MadTinkerer says:

    The important thing is that Valve have finally made an example out of Digital Homicide. DH are not the only shady jerks shovelling worthless technically-not-illegal incomplete buggy messes onto Greenlight, then cheating the voting process to take the spot that would otherwise go to a hard working Indie who spent years instead of weeks on their game. But DH are the loudest and most litigious.

    Good riddance.

  21. Rindan says:

    Review systems are universally shit for one reason. Fake/motivated reviews. Amazon is almost worthless at this point because of it. Steam is getting there in terms of indie games. Anything they can do to kill it now is good. Friendly fire is okay. I’d rather a game have no reviews than 20 “motivated” reviews.

    Yes, I know the developers most certainly feels differently, but in the long run it is even better for them. It is better to have a system where the best game wins with a dash of luck in terms of discoverability, than a system where a the indie level it is all who can market the best and manipulate their scores.

    Like I said though, in the end, better to error on killing motivated reviews and doing collateral damage, than to let motivated reviews foul up the entire review system.